A CLI tool and an API for fetching data from Twitter for free!
- NodeJS 20
- A working Twitter account (optional)
It is recommended to install the package globally. Use the following steps to install the package and ensure it's installed correctly:
- Open a terminal.
- Install the package using the command
npm install -g rettiwt-api
. - Check if the package is installed correctly using the command
rettiwt help
.
Rettiwt-API can be used with or without logging in to Twitter. As such, the two authentication strategies are:
-
'guest' authentication (without logging in) grants access to the following resources/actions:
- Tweet Details
- User Details
- User Timeline
- User Replies Timeline
-
'user' authentication (logging in) grants access to the following resources/actions:
- Tweet Details
- Tweet Like
- Tweet Likers
- Tweet List
- Tweet Post
- Tweet Retweet
- Tweet Retweeters
- Tweet Search
- Tweet Stream
- Tweet Unlike
- Tweet Unpost
- Tweet Unretweet
- Tweet Media Upload
- User Details
- User Follow
- User Followers
- User Following
- User Highlights
- User Likes
- User Media
- User Replies Timeline
- User Subscriptions
- User Timeline
- User Unfollow
By default, Rettiwt-API uses 'guest' authentication. If however, access to the full set of resources is required, 'user' authentication can be used, which requires the following additional steps post-installtion:
-
Open a terminal.
-
Generate an API_KEY using the command
rettiwt auth login "<email>" "<username>" "<password>"
Here,
<email>
is the email of the Twitter account to be used for authentication.<username>
is the username associated with the Twitter account.<password>
is the password to the Twitter account.
-
The string returned after running the command is the API_KEY. Store it in a secure place for later use.
The API_KEY generated by logging in is what allows Rettiwt-API to authenticate as a logged in user while interacting with the Twitter API ('user' authentication). As such it is a very sensitive information and therefore, must be stored securely. The following points must be kept in mind while using the API_KEY for 'user' authentication:
- The API_KEY is generated by logging into Twitter using the email, username and password and encoding the returned cookies as a base64 string. This encoded string is the API_KEY.
- The API_KEY provides the same level of authorization as any standard Twitter account, nothing more, nothing less.
- Since generation of API_KEY is equivalent to logging in to Twitter, repeated generation attempts might trigger Twitter's anti-bot measures, the same way repeated login attempts do.
- The API_KEY expires after one year from the day it was generated.
- Therefore, it is recommended to generate the API_KEY only once, then use it every time it is needed.
- Do not generate an API_KEY if it has not expired yet!
- If you have no idea of programming, it's recommended to use the CLI.
- The CLI provides an easy to use interface which does not require any knowledge of JavaScript or programming
- Please skip to CLI-Usage for details.
Rettiwt-API can be used as a dependency for your NodeJS project. In such a case, it is not required to install Rettiwt-API globally and you may install it locally in the root of your project using the command:
-
npm install --save rettiwt-api
(using npm)or
-
yarn add rettiwt-api
(using yarn)
However, in this case, for accessing the CLI, you will be required to prepend the CLI commands with npx
in order to tell NodeJS to use the locally installed package.
For example, for generating the API_KEY, the command will be modified as follows:
npx rettiwt auth login <email> <username> <password>
When used as a dependency, the Rettiwt class is entry point for accessing the Twitter API.
A new Rettiwt instance can be initialized using the following code snippets:
const rettiwt = new Rettiwt()
(for 'guest' authentication)const rettiwt = new Rettiwt({ apiKey: API_KEY })
(for 'user' authentication)
The Rettiwt class has three members:
auth
memeber, for managing authenticationtweet
member, for accessing resources related to tweetsuser
member, for accessing resources related to users
For details regarding usage of these members for accessing the Twitter API, refer to Features.
The following examples may help you to get started using the library:
import { Rettiwt } from 'rettiwt-api';
// Creating a new Rettiwt instance
// Note that for accessing user details, 'guest' authentication can be used
const rettiwt = new Rettiwt();
// Fetching the details of the user whose username is <username>
rettiwt.user.details('<username>')
.then(details => {
...
})
.catch(error => {
...
});
import { Rettiwt } from 'rettiwt-api';
// Creating a new Rettiwt instance using the API_KEY
const rettiwt = new Rettiwt({ apiKey: API_KEY });
/**
* Fetching the list of tweets that:
* - are made by a user with username <username>,
* - contain the words <word1> and <word2>
*/
rettiwt.tweet.search({
fromUsers: ['<username>'],
words: ['<word1>', '<word2>']
})
.then(data => {
...
})
.catch(err => {
...
});
For more information regarding the different available filter options, please refer to TweetFilter.
The previous example fetches the the list of tweets matching the given filter. Since no count is specified, in this case, a default of 20 such Tweets are fetched initially. The following example demonstrates how to use the cursor string obtained from the response object's next field, from the previous example, to fetch the next batch of tweets:
import { Rettiwt } from 'rettiwt-api';
// Creating a new Rettiwt instance using the API_KEY
const rettiwt = new Rettiwt({ apiKey: API_KEY });
/**
* Fetching the list of tweets that:
* - are made by a user with username <username>,
* - contain the words <word1> and <word2>
*
* 'data' is the response object received in the previous example.
*
* 'count' is a number less or equal to 20 (the quantity of tweets to return)
*/
rettiwt.tweet.search({
fromUsers: ['<username>'],
words: ['<word1>', '<word2>']
}, count, data.next.value)
.then(data => {
...
})
.catch(err => {
...
});
Sometimes, you might want to generate an API_KEY on the fly, in situations such as implementing Twitter login in your application. The following example demonstrates how to generate an API_KEY during runtime:
import { Rettiwt } from 'rettiwt-api';
// Creating a new Rettiwt instance
const rettiwt = new Rettiwt();
// Logging in an getting the API_KEY
rettiwt.auth.login('<email>', '<username>', '<password>')
.then(apiKey => {
// Use the API_KEY
...
})
.catch(err => {
console.log(err);
});
Where,
<email>
is the email associated with the Twitter account to be logged into.<username>
is the username associated with the Twitter account.<password>
is the password to the Twitter account.
For masking of IP address using a proxy server, use the following code snippet for instantiation of Rettiwt:
/**
* PROXY_URL is the URL or configuration for the proxy server you want to use.`
*/
const rettiwt = new Rettiwt({ apiKey: API_KEY, proxyUrl: PROXY_URL });
This creates a Rettiwt instance which uses the given proxy server for making requests to Twitter.
When using this library in an application deployed in a cloud environment, the library might throw error 429, even when under rate limits. This happens because Twitter's v1.1 API endpoints seemingly blocks access from cloud services' IP ranges. These v1.1 API endpoints are the ones used for authentication and as such, authentication tasks are blocked while deployed on cloud environments.
This issue can be bypassed by using a proxy only for authentication, using the following code snippet for creating a new Rettiwt instance:
const rettiwt = new Rettiwt({ authProxyUrl: PROXY_URL });
Where,
PROXY_URL
is the URL to the proxy server to use.
Authentication proxy is required only in the following two scenarios:
- While using 'guest' authentication.
- While creating API_KEY by 'user' authentication.
Sometimes, when the library shows unexpected behaviour, for troubleshooting purposes, debug logs can be enabled which will help in tracking down the issue and working on a potential fix. Currently, debug logs are printed to the console and are enabled by setting the 'logging' property of the config to true, while creating an instance of Rettiwt:
/**
* By default, is no value for 'logging' is supplied, logging is disabled.
*/
const rettiwt = new Rettiwt({ apiKey: API_KEY, logging: true });
Rettiwt-API also provides direct access to the raw response data, bypassing any preprocessing by the library itself. This can be achieved by using the FetcherService
class instead of the Rettiwt
class, as demonstrated by the example below, which fetches the raw details of a user with the username 'user1':
import { FetcherService, EResourceType } from 'rettiwt-api';
// Creating a new FetcherService instance
const fetcher = new FetcherService({ apiKey: API_KEY });
// Fetching the details of the given user
fetcher
.request(EResourceType.USER_DETAILS_BY_USERNAME, { id: 'user1' })
.then((res) => {
console.log(res);
})
.catch((err) => {
console.log(err);
});
import { FetcherService, EResourceType, IUserDetailsResponse } from 'rettiwt-api';
// Creating a new FetcherService instance
const fetcher = new FetcherService({ apiKey: API_KEY });
// Fetching the details of the given user
fetcher
.request<IUserDetailsResponse>(EResourceType.USER_DETAILS_BY_USERNAME, { id: 'user1' })
.then((res) => {
console.log(res);
})
.catch((err) => {
console.log(err);
});
As demonstrated by the example, the raw data can be accessed by using the request
method of the FetcherService
class, which takes two parameters. The first parameter is the name of the requested resource, while the second is an object specifying the associated arguments required for the given resource. The complete list of resource type can be checked here. As for the resource specific argurments, they are the same as that of the methods of Rettiwt
class' methods for the respective resources, but structured as an object. Notice how the FetcherService
class takes the same arguments as the Rettiwt
class, and the arguments have the same effects as they have in case of Rettiwt
class.
So far, the following operations are supported:
- Getting the details of a tweet
- Liking a tweet
- Getting the list of users who liked a given tweet
- Getting the list of tweets from a given Twitter list
- Posting a new tweet
- Retweeting a tweet
- Getting the list of users who retweeted a given tweet
- Searching for the list of tweets that match a given filter
- Streaming filtered tweets in pseudo-realtime
- Unliking a tweet
- Unposting a tweet
- Unretweeting a tweet
- Uploading a media file for a tweet
- Getting the details of a user
- Following a given user
- Getting the list of users who follow the given user
- Getting the list of users who are followed by the given user
- Getting the list of highlighted tweets of the given user
- Getting the list of tweets liked by the given user
- Getting the media timeline of the given user
- Getting the replies timeline of the given user
- Getting the list of subscriptions of the given user
- Getting the tweet timeline of the given user
- Unfollowing a given user
Rettiwt-API provides an easy to use command-line interface which does not require any programming knowledge.
By default, the CLI operates in 'guest' authentication. If you want to use 'user' authentication:
- Generate an API_KEY as described in Authentication.
- Store the output API_KEY as an environment variable with the name 'API_KEY'.
- Additionally, store the API_KEY in a file for later use.
- Make sure to generate an API_KEY only once, and use it every time you need it.
- The CLI automatically reads this environment variable to authenticate against Twitter.
- Additionally, the API_KEY can also be passed in manually using the '-k' option as follows:
rettiwt -k <API_KEY> <command>
- Additionally, the API_KEY can also be passed in manually using the '-k' option as follows:
Help for the CLI can be obtained from the CLI itself:
- For help regarding the available commands, use the command
rettiwt help
- For help regarding a specific command, use the command
rettiwt help <command_name>
The complete API reference can be found at this page.
- This API uses the cookies of a Twitter account to fetch data from Twitter and as such, there is always a chance (although a measly one) of getting the account banned by Twitter algorithm.
- There have been no reports of accounts getting banned, but you have been warned, even though the chances of getting banned is negligible, it is not zero!